用生存分析控制生成长度，提速7倍且不丢精度。
Survival-Guided Length Control for Efficient Diffusion Language Models
- 将序列结束预测建模为生存问题，动态决定生成长度。
- 在推理中实现最高7倍加速，任务准确率不变。
- 适合追求高效生成的开发者与研究者使用。
扩散语言模型通过迭代去噪掩码序列生成文本，但标准解码要么固定序列长度，要么依赖随意的停止规则，常导致不必要的去噪步骤。本文将长度选择重新建模为对序列结束标记的离散时间生存问题，提出一种即插即用、无需训练的长度预测器，可集成到任意现有扩散语言模型中。在推理任务和代码生成基准上，基于生存引导的长度解码使推理速度提升至多7倍，同时保持任务准确性。进一步发现，即使在同一数据集内，预测长度也存在显著差异，表明模型性能对所选长度敏感。
原文摘要 · Abstract (English)
Diffusion language models (DLMs) generate text by iteratively denoising masked sequences, but standard decoding either fixes the sequence length or relies on ad hoc stopping rules, often leading to unnecessary denoising steps. We recast length selection as a discrete-time survival problem over the end-of-sequence token and propose a plug-in, training-free length predictor that can be added to any existing DLM. Across reasoning and code-generation benchmarks, survival-guided length decoding speeds up inference by up to 7 times while preserving task accuracy. We further find that predicted lengths vary widely even within the same dataset, making model performance sensitive to the chosen length.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。